This new competing continuation application seeks a further five years support for a methodology and data-analysis project focused on the development and testing of genetic models for the natural history of alcohol use and abuse across the life-span. Computer simulation and theoretical analysis will be used to explore four related areas of alcoholism research: (i) high-risk research paradigms, including their power and predictions under different genetic models, and their strengths, limitations and alternatives; (ii) inferences about causality, and the problem of resolving reciprocal and indirect causation hypotheses; (iii) multi-stage genetic models for the natural history of substance use patterns and substance use disorders; and (iv) social environmental influences on patterns of alcohol use and alcoholism risk, and their incorporation into genetic and high-risk research (including models for GE correlation and GxE interaction). Attention will be focused on ways in which genetic, high-risk research and social environmental research traditions can be integrated, and on the identification of research strategies which best allow such an integration; and on the implications of such complications as etiologic heterogeneity, psychiatric comorbidity etc. for genetic modelling. Illustrative data analyses will use data on the histories of alcohol and tobacco use, and associated behavioral risk- factors, of adult twins and their, adult relatives (parents, spouses, siblings and children: N=50,000) from twin-family studies in Australia and in the U.S. (including two-wave prospective data on 2900 adult Australian twin pairs assessed in 1981 and 1989). Data-analyses will focus on (i) the interaction of genetic differences and changing social environment in determining the shift from the unstable drinking patterns of early adulthood to the more highly stable and highly familial patterns of adulthood; (ii) the role of assortative mating and reciprocal spousal environmental influence in determining spousal concordance for substance use patterns, and its implications for the familial aggregation of substance use patterns; (iii) intergenerational analyses of the transmission of substance use patterns, and postulated mediating personality, attitudinal and sociodemographic variables, to address issues of the intergenerational continuity of genetic influence on substance use (G x cohort interaction).